November 29, 2016
The Rise and Fall of the Power Rankings
If you have a moment, join me as we hop onto the Internet Archive Wayback Machine. Set your watch to that exciting, lawless day of our generation’s lone frontier: the late nineties. The place: a humble website, spelled out awkwardly by many a suave sports anchor in the twilight moments of a late-night SportsCenter: espnet dot sportszone dot com.
Gawk at the unkempt wilds of Internet 1.0, an unconquered continent where bylines were rare, jpegs were a bandwidth luxury, and statistics were still halfway from GWRBI to GB%. It’s an interesting glimpse into the culture of sports internet. (One link ahead of its time: ESPN’s hosted video game review site.) Familiar names dot the page: There’s Neyer! There’s Caple! There’s Sickels! There’s Kurkjian!
But missing are not only the images and logos that dot the modern website front page, the embedded videos and tweets and advertisements, but the headshots. The idea of author blogs, or video interviews, is far off. While these are respected writers now, and were respected in the industry then, they weren’t names yet: ESPN, long powered on the television side by the power of personality, hadn’t yet finished crafting their writers into authorities.
Perhaps less noticeably absent, at first, is the absence of a prominent feature in later iterations of espn.go.com: the Power Ranking. ESPN’s foray into the ranking business, dubbed the Power Alley, arrived in 2002. (Click that link, if only to enjoy the first sentence.) The format hasn’t really changed in 14 years: Team logo and name, team record, a 1-2 sentence stinger to explain their positioning or supply a fun fact. Repeat 29 times. The work is simple, and the response to stimuli is undeniable; it takes a long, generally predictable six months of regular season and splits it into little championships, little seasons of their own. It provides progress, but most of all, it provides validation.
Princeton Sports Analytics recently performed a review of the accuracy of ESPN’s power rankings, going back six years and comparing the final rank of each team for each of the major four sports with the first week. Author Ben Ulene found that for baseball, the correlation returned an R-squared value of .24, putting it behind basketball and hockey, and ahead of football (although football also outpaced baseball in four of the six seasons). He posits that roster size, high scores, and injury rates contribute to a league’s volatility. Outside the reach of the study, but vitally important to most fans, is what happens after that final week, where the madness that is the MLB playoffs further throws the power of prediction into doubt.
All power rankings are essentially annotated standings; their original function was to employ expertise to distinguish between performers who faced unbalanced schedules (college football and basketball) or who did not play often enough to set a baseline (tennis, mixed martial arts). It’s the simplest form of value-added content: facts already available to the common reader, laced with grade-A expertise.
But while the nerds were busy taking over the front offices in baseball and winning championships, they were changing the way we rooted for those championships, too. It’s not that the statistical revolution has killed expert analysis; if anything, it’s exploded the scene (along with fantasy sports) to the point where everyone is (or has the tools to be) an expert. And that’s the transition; the expansion and publicization of statistics has made analysis more transparent, more verifiable than ever before. This is, beyond doubt, a good thing, not just for those of us able to do the verifying, but the experts themselves. No longer do two people have to battle each other in front of us for ultimate renown, but each person can battle the educated public him or herself. There has never been more conversation.
The older mediums of television and radio remain necessarily top-down, but the internet allows not only feedback but time to craft it. That reciprocal relationship between author and reader, expert and layman, so embedded into the culture of the modern internet has rendered the power ranking a strange vestigial tail. It’s the Hall of Fame in a world of WAR leaderboards, except somehow even more extraneous and forgettable. They’ve become so perfunctory that the most interesting facet to them is no longer the rankings, but the power: like the MVP discussion, it’s now the question, the definition of the assertion, that stands out, rather than the answers. We now have fielding-independent statistics and Statcast data to derive talent from performance; we have playoff odds that extract championships, even championship percentage added, from talent.
Each power ranking has to decide what to be in this brave new world. Is it sorting by wins and then adjusting for the invisible hand of momentum? Is it a strict ordering of World Series victory odds, dictated by the favored local projection system? Is it derived from some obscure mysticism, chicken bones rattled in the cup of an anonymous author? Each of these is fine, assuming they admit to their foundations. What ultimately sets one ranking apart from another, given how little predictive value they provide on average, is the quality of the writing. That little stinger after each team, once filler meant to convey the valuable ranking, now works in the opposite direction. Now, as is the case for BP’s own Hit List, the odds are an excuse to present 30 excellent one-liners, while still providing a quick, unscientific map of the league at that moment.
But we’ve always known the odds, if not necessarily to the third decimal point; they’re intuitive, a barometer of hope. We internalize our power rankings, misjudge, and are roughly reminded by the trade deadline fire sale or the countdown of magic numbers in a lost September or a called third strike in a divisional playoff game. There’s always a moment when the hope and the hype of the power ranking cleaves, violently, with reality. It’s the in between where this creation exists, the waiting for judgment, the mornings and the off days when there is nothing to do but talk.
And this is the allure: that the power ranking quantifies that shiniest of sports pyrite, the respect of the national media. The prize is the attention of the authorities, their father-like approval. It’s that paternalism in sports media that readers are, hopefully, moving past.